Title :
Blind Source Separation Using Decoupled Relative Newton Algorithm
Author_Institution :
Fortemedia Inc., Sunnyvale, CA, USA
Abstract :
A decoupled relative Newton algorithm is proposed for the matrix optimization problem encountered in blind source separation (BSS) and independent component analysis (ICA). The algorithm decouples the matrix optimization problem into a series of small vector optimization problems. The nonsingularity of separation matrix enables a simple and efficient relative Newton learning algorithm for the vector optimization problems. Simulation results are reported to confirm its superior performance.
Keywords :
Newton method; blind source separation; independent component analysis; learning (artificial intelligence); matrix algebra; optimisation; BSS; ICA; blind source separation; decoupled relative Newton algorithm; independent component analysis; matrix optimization problem; relative Newton learning algorithm; separation matrix nonsingularity; vector optimization problems; Algorithm design and analysis; Blind source separation; Convergence; Optimization; Signal processing algorithms; Vectors; Blind source separation; Newton algorithm; independent component analysis; relative gradient;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2012.2207890
